In 2 experiments, the effects of predictive inference level and text process depth on predictive inference generation were tested. And the applicability of the go/no-go lexical decision task in text comprehension studies was also tested. 2 × 2 independent design , response time index and longer text material were used in both experiment 1 and experiment 2 which separately used yes/no and go/no-go lexical decision task. In experiment 1, there were significant main effects of both predictive inference level and text process depth, but a nonsignificant interaction between predictive inference level and text process depth. In experiment 2, there wasn't significant main effect of predictive inference level, while there was a significant main effect of text process depth and a nonsignificant interaction between predictive inference level and text process depth. The only different aspect between the results of the 2 experiments was on the main effect of predictive inference level. Through analysis, the result of experiment 1 was accepted. The eventual result is that: predictive inference level has effect on the on-going generation of predictive inference content; text process depth has effect on the on-going generation of predictive inference content; The effects of predictive inference level and text process depth on the on-going generation of predictive inference are independent, and the two effects can be added up; Go/no-go lexical decision task can be used in the study field of text comprehension, but it's sensitivity is less than yes/no lexical decision task.
|